A facile route to the [4.1. 1] propellane system

Jun 7, 1978 - data points falling very close to a straight line. ... these uncertainties our results lead to the unambiguous con- ... Shy-Fuh Lee, Cii...
0 downloads 0 Views 284KB Size
3966

Journal of the American Chemical Society

the temperature. Using relation 1, C, is calculated a t each temperature from which the equilibrium constant K and the Gibbs free energy AGO can be obtained (Table I). The enthalpies AH" were obtained from the Arrhenius plots, the data points falling very close to a straight line. Aside from experimental uncertainties the accuracy of the calculated thermodynamic data is dependent on the accuracy of the assumed rotational strengths for the conformers involved in the equilibrium. Any uncertainty of the deuterium contribution to [RIax and [RIeq will strongly affect C, K , and AGO, whereas any uncertainty in the methyl contribution will be reflected predominantly in the magnitude of A H " . Based on the extensive experimental material, the estimates for the methyl groupsIo are probably good within 10%. Estimates for the deuterium contribution^^^^ are based on the rigid adamantanone ring system, and whether these are ideal representatives for the more flexible cyclohexanone system must be left open until more data become available. Irrespective of these uncertainties our results lead to the unambiguous conclusion that axially oriented deuterium is energetically preferred oCer the corresponding equatorial one. A full discussion of the implications of this work together with experimental details will appear in our full paper.

Acknowledgment. Partial financial support by the National Institute of Health (Grant No. G M 20276) and experimental assistance by Ruth Records are gratefully acknowledged. References and Notes (1) For part 122 in this series, see ref 6. (2) (a) W. C. M. C. Kokke and L. J. Oesterhoff, J. Am. Chem. Soc., 94,7583 (1972); (b) W. C. M. C. Kokke and L. J. Oesterhoff, ibid., 95, 7159 (1973). (3) R. F. R. Dezentje and H. P. J. M. Dekkers, J. Chem. Phys., 18, 189 ( 1976). (4) D. A. Lightner, T. C. Chang, and J. Horwitz, Tetrahedron Lett., No. 35, 3019 (1977). (5) H. Numan and H. Wynberg, J. Org. Chem., in press. (6) C. Djerassi, C. L. VanAntwerp, and P. Sundararaman, Tetrahedron Lett., 535 (1978). (7) J. W. Simek, D. L. Mattern, and C. Djerassi, Tetrahedron Lett, 3671 (1975). (8) P. H. Hoffman, E. C. Ong, 0. E. Weigang, Jr., and M. J. Nugent, J. Am. Chem. SOC.,96,2620 (1974). (9) (a) G.Snatzke and G. Eckhard, Tetrahedron, 24,4543 (1968); (b) G. Snatzke, B. Ehring, and H. Klein, ibid., 25, 5601 (1969). (10)D. N. Kirk and W. Kiyne, J. Chem. SOC.,Perkin Trans. 1, 1076 (1974). (11) The values for R and [RIeqgiven in Figure 1 were estimated from literature ~alues!~~~'8'8ssumingthe chair conformation for both conformers and additivity of the contributions from each perturber (CY,, methyl, 5.31 deuterium, (con): oeqmethyl, 0.66 (con); $eq deuterium, 0.335 (dis); 0.074 (dis)). Values for the methyl contribution were converted from Ac to [R] using the empirical relation [R] = 3.32 X Ac; the [R] values for p deuterium were obtained by direct integration of the given spectra (con = consignate, dis = dissignate). (12) H. Kosrnol, K. Kieslich, R. Vossing, H. J. Koch, K. Petzold. and H. Gibian, Justus Liebigs Ann. Chem., 701, 198 (1976), have shown that such reductions provide the S alcohol. (13) H. J. Liu and S. P. Majumdar, Synth. Commun., 5, 125 (1975). (14)A. P. Krapcho and A. J. Lovey, Tetrahedron Lett., 957 (1973). (15) R . E. ireiand and J. A . Marshall, J. Am. Chem. SOC.,81, 6336 (1959). (16) J. C. Fairly, G. L. Hodgson, and T. Money, J. Chem. SOC.,Perkin Trans I , 2109 (1973). (17) Pharma Forschung, Schering, AG., DlOOO Berlin 65, West Germany, where all microbiological reductions were performed.

a,,

Department of Chemistry, Stanford University Stanford, California 94305 Receiced March 9, 1978

A Facile Route to the [4.1.1]Propellane System Sir: Recently we presented arguments for the formation of tricyclo[4.1 .0.02s7]hept-1(7)-ene ( 2 ) as a reaction intermediate in the nucleophilic substitution of 1-chlorotri0002-7863/78/1500-3966%01.OO/O

100:12

/

June 7, 1978 li

3

cyclo[4.1 .0.02,']heptane (1) with organolithium compounds leading to the corresponding I-alkyl- or 1aryltricyclo[4.1 .0.02,7]heptanes.' These results initiated an attempt to trap the proposed intermediate 2 with a 1,3-diene in a Diels-Alder reaction. The outcome of this experiment is reported in this communication. When l 2in THF was slowly added to a stirred suspension of anthracene (2 equiv) and lithium 2,2,6,6-tetramethylpiperidide (1.5 equiv) in T H F at -20 "C, aqueous workup afforded 9,10-dihydro-9,10-( 1,7-tricyclo[4.1 .0.02,7]heptane)anthracene (3) in an isolated yield of 3 1% (mp 160- 162.5 "C, from n - ~ e n t a n e ) . ~ Structure proof for 3 is based on the mass spectrum (m/e 270 (M+)), the 'H N M R spectrum ((CDC13) 6 1.26 (narrow m, 6 H, 13-H2, 14-H2, 15-H2), 2.18 (narrow m, 2 H, 12-H, 16-H), 4.60 (s, 2 H , 9-H, 10-H),7.15 (AA'BB'system, 8 H, aromatic protons)), and the I3C N M R spectrum ((CDC13) 6 20.16 (t, C-14), 22.41 (t, C-13, C-15), 22.55 (s, C-11, C-17), 46.68 (d, C-9, C-IO), 53.91 (d,C-12,C-16), 124.78, 125.91 (d, C-1.C-4, C-5, C-8, and C-2, C-3, C-6, C-7, or reversed), 141.21 (s, C-4a, C-8a, C-9a, C-10a)).4 The chemistry of 3 is in accord with the bicyclo[ 1 . I .O]butane structure. In the temperature range of 160- 180 " C 3 cleanly rearranged to 9, IO-dihydro-9,lO-[2,7-(3-methylenecyclohex-I-eno)]anthracene ( 5 ) : mp 21 1-213 "C, from pentane: mass spectrum m/e 270 ( M f ) ; IH N M R (CDC13) 6 1.15-1.71 (m, 2 H), 1.82-2.23 (m, 4 H), 4.23 (d, J = 8.5 Hz, 1 H), 4.46 (s, 1 H), 5.82-6.03 (m, 2 H), 6.95-7.37 (m, 8 H); I3C N M R (CDC13) 6 22.29 (t), 26.25 (t), 32.61 (t), 47.52 (d), 57.90 (d), 123.78 (d), 124.38 (d), 124.81 (d), 125.36 (d), 125.57 (d, unresolved superposition of two signals), 13 1.22 (s), 134.99 (s), 140.61 (s), 144.20 (s). The probable mechanism for this conversion is a retro-carbene ring opening of the bicyclo[ 1.1 .O] butane unit in 3 leading to the intermediate 4 which is then transformed to 5 by hydrogen migration from C- 12 to the carbene enter.^

L

J

4

Shy-Fuh Lee, Ciinter Barth Klaus Kie~lich,~' Carl Djerassi*

/

-5

The diene 5 was also produced in a vigorous reaction when 3 was mixed with AgBF4 in CgD6.637 The central bond C- 1 1-C- 17 of the bicyclo[ I . 1 .O] butane system in 3 was cleaved selectively when thiophenol was combined with 3 and allowed to add via a radical-chain process.8 9,1O-Dihydro-9,10- [6-exo, 7-anti-(6-endo-phenyIthio)norpinano]anthracene ( 6 ) was isolated in a practically quantitative yield: mp 197-198 "C, from ethanol: mass spectrum m/e 380 (M+): IH N M R (CDC13) 6 1.43-2.37 (br m, 8 H), 2.63 and 4.17 (AB system, J = 9.5 Hz), 4.44 (s, 1 H), 6.63-7.22 (m, 8 H), 7.27 (br m, 5 H); I3C N M R (CDC13) 6 14.45 (t), 26.92 (t), 37.96 (d), 47.61 (d),47.85 (d), 57.82 (s), 58.24 (d), 124.16 (d), 125.34 (d), 125.53 (d), 125.68 (d),

0 1978 American Chemical Society

3967

Communications to the Editor

Q

Metal Ion Binding to Cytidine in Solution. Compelling Raman and Carbon-13 Nuclear Magnetic Resonance Spectral Evidence for Coordination to the Exocyclic Oxygen at Position 2

PhS

126.61 (d), 128.65 (d), 132.33 (d), 134.66 (s), 143.25 (s), 144.70 (s). It is interesting to note that the adduct 3 contains the structural unit of a [4.l.l]pr0peIlane.~This system is now easily accessable via the procedure outlined above. Judging from molecular models, the bicyclobutane bridgehead carbon atoms C-l 1 and C-17 in 3 are positioned outside the tetrahedron formed by the four corresponding substituents, C- 10, C-12, C-16, C-17 for C-11, and C-9, C-11, (2-12, C-16 for c - 17.I0,l I The formation of the Diels-Alder adduct 3 can be regarded as unequivocal proof for the existence of 2 as an intermediate only, if one were able to exclude alternative routes from 1 to 3 under the experimental conditions employed. This is, however, not the case. The bridgehead bicyclobutyl anion derived from 1 might add to anthracene, the intermediate 7, after

&& /

/ \

\ / '

", \

\

H

H

7

8

i

electron transfer from the carbanionic center to the halogen, might lose the chloride anion, and might form the diradical8 that could close the ring. Experiments to differentiate between these mechanistic possibilities are in progress.

Acknowledgment. This work was supported by Deutsche Forschungsgemeinschaft and by Fonds der Chemischen Industrie. References and Notes (1) G. Szeimies, J. Harnisch, and 0. BaumgArtel, J. Am. Chem. Soc., 99, 5183 (1977). (2) G. Szeimies, F. Philipp, 0. Baumgartel, and J. Harnisch, Tetrahedron Lett., 2135 (1977). (3) The yield was not optimized. (4) The values of combustion analysis and of osmometric molecular weight determination are consistent with C2,Hq8. (5) For a recent example of a thermal retro-carbene ring opening of a bicyclo[l.l.O]butane system, see G. Szeimies, J. Harnisch, and K.-H. Stadler, Tetrahedron Lett., 243 (1978). (6) For a recent review on transition metal ion catalized isomerizations of strained hydrocarbons, see K. C. Bishop ill, Chem. Rev., 76, 461 (1976). (7) In addition to 5 a second, not yet identified, isomer was formed in 11YO yield. (8) G. Szeimies, A. SchloBer, F. Philipp, P. Dietz, and W. Mickler, Chem. Ber., 111, 1922 (1978). (9) D. Ginsburg, "Propellanes: Structure and Reactions", Verlag Chemie, WeinheimIBergstr., 1975. (10) For examples of molecules with "inverted tetrahedron", see K. 8. Wiberg, G. J. Burgmaier. K. W. Shen. S. J. La Placa, W. C. Hamilton, and M. D. Newton, J. Am. Chem. SOC.,94, 7402 (1972). and references therein. (1 1 ) After submission of this manuscript the x-ray structure of 3 has been determined by J.-P. Declercq, G. Germain, and M. Van Meerssche (University of Louvain, Belgium) which confirms the proposed structure for 3 and the "inverted tetrahedron" for C-11 and C-17. We thank Professor Van Meerssche for communicating these results to us prior to publication.

Ursula Szeimies-Seebach, Giinter Szeimies* Institut fiir Organische Chemie, Uniuersitat Miinchen Karlstrasse 23, 0-8000 Miinchen 2, West Germany Receiced January 23, 1978

0002-7863/78/1500-3967$01.00/0

Sir: Diverse electrophiles (H+, Pt", Hgl1 compounds), which bind to N(3) of cytosine derivatives,] cause characteristic Raman difference spectra in the 1100-1400-cm-~ region in a q u e o u ~and ~ ? dimethyl ~ s ~ l f o x i d e - d 6solutions. ~ We observe a new type of difference spectrum which can be attributed to the binding of hard metal ions to O(2) of cytidine in Me2SOd6.Definitive evidence for O(2) binding in solution is lacking but such binding occurs frequently in The problem in distinguishing O(2) and N(3) binding sites in solution can be appreciated when it is realized that the sites are strongly coupled electronically and that the expected metal-O(2)-C(2) angle8 places the metal ion a t a position close to that expected for a n N(3) bound metal atom. Except for uridine (thymidine) and N ( 1) of guanosine, similar complexes are formed between metal ions and nucleosides in water and in M e 2 S 0 . l ~The ~ high solubilities obtainable in Me2SO permit weak interactions to be observed. The existence of such interactions for alkaline earth metal ions is the subject of both intense interest and controversy.I~Io-12 For example, the nature of the interaction of Mg2+ with the adenine ring in the Mg2+-ATP complex is unclear.13 We showed that the-I-ppm shifts observed in the ' H N M R spectrum of guanosine (in Me2SO) upon addition of alkaline earth chloride salts1l were largely due to hydrogen-bonding interactions with the chloride counterion.I2 A large chloride ion effect is also observed for ~ y t i d i n e . ' ~The ? ~ relatively ~,~~ small but reproducible I3C shifts observed for cytidine in Me2SO-d6 on addition of alkaline earth chloride saltslo cannot be attributed to the chloride ion effect.I0.l6 In general, the changes in shifts of ligand resonances induced by metal coordination are difficult to interpret. Thus, it was reportedt0that alkaline earth salts caused downfield shifts in the C(2) resonance of cytidine in MezSO-d6, whereas HgClz which binds to N(3) causes large upfield shifts and ZnClz (also believed to bind at N(3))17 causes small upfield shifts of this resonance. Most metal species induce upfield shifts of C(4), but the alkaline earth cations were reported to have a diversity of effects.10x18Electrophiles known to bind to N(3) cause upfield shifts in both the C(2) and the C(4) of cytidine in MezS0I6 and aqueous solutions.2 Results that we have obtained with nitrate salts are given in Table I. Raman difference spectroscopy is much less subject than N M R spectroscopy to environmental effects.'-3 The Raman results have been readily interpretable because diverse N (3) bound electrophiles, such as HgC12, H+, and Ptl1complexes, cause similar difference spectra.'-3 The effects of HgClz and H+ (not shown) on the Raman spectrum of cytidine are similar Table I. Metal Ion Induced I3C N M R Shifts in Cytidine" Salt ( M )

HgClz (0.5) Zn(NO3)>(0.5) Ba(NO& ( O S ) La(N03)3 (0.47) Pr(NO3)3 (0.46) Corrected

C-2

C-4

C-5

C-6

C-I'

+2.7 C0.2 -0.8 -0.8

+2.2

-1.7 -1.3 -0.6 -1.0

-0.6 $0.1 t0.3 +0.5

-1.4

-0.9 -0.6 -0.1 -0.1 $0.7

-0.4

f0.8

$0.6

+10.1

+10.9

fl.2 +OS fO.5 +2.1 +1.6

$1.1

a 0.2 M in MezSO-ds, in parts per million, Me4Si reference, upfield shifts from free ligand values positive, Varian CFT-20 instrument. Shifts for the metal free solutions were as follows: C-2 (155.4), C-4 (165.5), C-5 (93.8), C-6 (l41.4),and C-l'(89.2). See ref 16 for assignments.

0 1978 American Chemical Society